PMA : des interventions complémentaires peu convaincantes

Par Mme Aude Rambaud
(Saint-Germain-en-Laye)

[Déclaration de liens d’intérêts]

 Date de publication : 17 juillet 2026 https://www.mediscoop.net/index.php?pageID=8608c95e7311a6e9e91a1b9c87951dd9&site_origine=newsletter_mediscoop&nuid=44baf5968540a6248a8065e80f2f7273&midn=24088&from=newsletter&slnk=4

emps de lecture: 2 min

---

La plupart des interventions complémentaires proposées aux femmes engagées dans un parcours de procréation médicalement assistée pour augmenter les chances de réussite, sont peu ou pas efficaces, avec des niveaux de preuve souvent insuffisants. C’est ce qu’indique une revue systématique et méta-analyse, parue dans The Lancet Obstetrics, Gynaecology & Women’s Health.

Pour améliorer les résultats de la procréation médicalement assistée, des traitements complémentaires sont proposés aux patientes : procédures, médicaments ou encore médecines alternatives. 

Une méta-revue présente des revues systématiques portant sur dix traitements complémentaires fréquemment recommandés en pratique :

-diagnostic génétique préimplantatoire pour l’aneuploïdie,

-test de réceptivité endométriale,

-corticoïdes,

-l’utilisation d’EmbryoGlue solution à base d’acide hyaluronique et d’albumine, –

-grattage endométrial,

-injection intracytoplasmique de spermatozoïdes (ICSI),

-activation ovarienne avec plasma riche en plaquettes,

-perfusion de plasma riche en plaquettes dans l’utérus,

-acupuncture,

-et perfusion intralipidique. 

Les essais inclus comparaient un ou plusieurs de ces traitements à une absence de traitement ou à un placebo. Le critère principal était la naissance d’un enfant vivant. 

Au total, 85 essais ont été inclus et 72 autres potentiellement éligibles, ont été exclus en raison d’un manque de fiabilité. L’analyse indique que le grattage de l’endomètre pourrait être associé à une légère augmentation des chances de naissance vivante (OR 1,20 ; p=0,02). 
Quatre interventions complémentaires ne semblaient avoir aucun effet : EmbryoGlue (OR 1,12 ; p=0,29), test Génétique Préimplantatoire (OR 1,16 ; p=0,23), test de réceptivité endométriale, et corticoïdes (OR 0,95 ; p=0,71). 

Pour cinq autres interventions, les données n’étaient pas concluantes : injection intracytoplasmique de spermatozoïdes (OR 1,12 ; p=0,16), acupuncture, perfusion intralipidique, activation ovarienne par injection de plasma et perfusion de plasma dans l’utérus.

Référence :
Lensen S et al.
Safety and effectiveness of ten common in-vitro fertilisation add-ons: a systematic review and meta-analysis.
The Lancet Obstetrics, Gynaecology, & Women’s Health, 23 juin 2026
Retrouvez l’abstract en ligne

Download PDF

https://www.thelancet.com/journals/lanogw/article/PIIS3050-5038(26)00054-3/fulltext

Summary

Background

Over the past decade, a growing number of in-vitro fertilisation (IVF) add-on therapies have become available. These IVF therapies are additional procedures, medications, or techniques intended to improve IVF success rates. Existing systematic reviews are outdated and many include studies with trustworthiness concerns. This meta-review provides comprehensive systematic reviews for ten common IVF add-ons, including the use of a trustworthiness assessment to identify and exclude untrustworthy trials.

Methods

In this systematic review and meta-analysis, we searched MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials from inception up to Jan 13, 2026 for prospectively registered randomised trials that had no serious trustworthiness concerns per the Trustworthiness in Randomised Controlled Trials (TRACT) checklist and compared one or more of the following IVF add-ons with no treatment or placebo in people using IVF: preimplantation genetic testing for aneuploidy (PGT-A); endometrial receptivity testing; corticosteroids; EmbryoGlue; endometrial scratching; physiological intracytoplasmic sperm injection; platelet-rich-plasma injection into the ovary; platelet-rich-plasma infusion into the uterus; acupuncture; and intralipid infusion. The primary outcome was live birth. Studies were evaluated using the Cochrane Risk of Bias 1 tool, and we used the TRACT checklist to identify trustworthiness concerns. The quality of evidence was assessed using the GRADE approach. Two reviewers independently screened and extracted data; disagreements were resolved by discussion. We did meta-analyses with the random-effects method and summarised the effects of the intervention as odds ratios (ORs) and 95% CIs. The protocol of each systematic review was prospectively registered (CD009517, CD007421, CD005996, CD009602, CD005291, CD005291, CD013875, CD006920, CD016209, and CD010461).

Findings

Of 12 398 search results, 85 trials were included across the ten systematic reviews and meta-analyses; 72 other potentially eligible studies were excluded for possible trustworthiness concerns, such as the absence of prospective trial registration. For the primary outcome of live birth, endometrial scratching might be associated with a small increase in the chance of live birth (OR 1·20, 95% CI 1·02–1·41; p=0·02; I²=39%; 19 randomised controlled trials [RCTs]; moderate certainty). Four add-ons appeared to have no effect: EmbryoGlue (OR 1·12, 0·91–1·37; p=0·29; I²=47%; seven RCTs; low-certainty evidence); PGT-A (OR 1·16, 0·91–1·47; p=0·23; I²=42%; five trials; moderate certainty); endometrial receptivity testing (three trials; meta-analysis not undertaken; moderate certainty); and corticosteroids (OR 0·95, 0·70–1·27; p=0·71; I²=0%; two trials; moderate certainty). For five add-ons the evidence was unclear because of the poor quality or scarcity of the data: physiological intracytoplasmic sperm injection (OR 1·12, 0·98–1·29; p=0·16; one trial; low certainty); acupuncture (ten trials; very-low certainty); and intralipid (one trial; very-low certainty); platelet-rich plasma injection into the ovary (one trial; very-low certainty); and platelet-rich plasma infusion into the uterus (two trials; very-low certainty).

Interpretation

In this systematic review and meta-analysis we found that most add-ons are not proven to benefit patients going through IVF. In many cases, the quality of evidence was poor, contributing to uncertainty regarding effectiveness, and many studies were excluded for trustworthiness concerns. Well designed, large-scale RCTs are needed to generate reliable evidence on the effect of these interventions.

Funding

The University of Melbourne and the Australian National Health and Medical Research Council.